Travel services based on transportation criteria

A system comprising:

• a memory that stores instructions;
• and
• one or more processors on a server configured by the instructions to perform operations comprising:
• determining a geographical location of a first user of a subscription-based travel service;
• searching a list of travel services that are available for the first user to consume on a travel date to identify candidate travel services;
• training a machine learning technique by processing training data comprising travel data associated with one or more users that include at least a given user, the training being performed by:
• retrieving a portion of the training data from a storage device;
• extracting travel activities features from the training data for the one or more users that include at least the given user;
• utilizing the machine learning technique to estimate a likelihood of consumption value for the given user based on the extracted travel activities features and based on one or more transportation criteria associated with arriving at a given travel service in the list of travel services from the geographical location of the given user;
• and
• updating parameters of the machine learning technique based on the estimated likelihood of consumption value for the given user;
• applying the machine learning technique to travel activities for the first user to estimate a likelihood of consumption value for the first user;
• determining that the estimated likelihood of consumption value for the first user is less than a threshold;
• determining first and second maximum purchase amounts based on first and second subscription values, the first subscription value corresponding to a first set of transportation criteria and the second subscription value corresponding to a second set of transportation criteria;
• increasing a travel cost value associated with the given travel service in response to determining that the likelihood of consumption value for the first user is less than the threshold, the travel cost value being based on at least one of the first or second maximum purchase amounts;
• and
• generating for display in a graphical user interface to the first user, an interactive visual representation of the given travel service that represents the travel cost value, the given travel service being associated with a cost that is greater than the first maximum purchase amount and less than the second maximum purchase amount.

A non-transitory computer-readable medium comprising instructions stored thereon that are executable by at least one processor to cause a computing device to perform operations comprising:

• determining a geographical location of a first user of a subscription-based travel service;
• searching a list of travel services that are available for the first user to consume on a travel date to identify candidate travel services;
• training a machine learning technique by processing training data comprising travel data associated with one or more users that include at least a given user, the training being performed by:
• retrieving a portion of the training data from a storage device;
• extracting travel activities features from the training data for the one or more users that include at least the given user;
• utilizing the machine learning technique to estimate a likelihood of consumption value for the given user based on the extracted travel activities features and based on one or more transportation criteria associated with arriving at a given travel service in the list of travel services from the geographical location of the given user;
• and
• updating parameters of the machine learning technique based on the estimated likelihood of consumption value for the given user;
• applying the machine learning technique to travel activities for the first user to estimate a likelihood of consumption value for the first user;
• determining that the estimated likelihood of consumption value for the first user is less than a threshold;
• determining first and second maximum purchase amounts based on first and second subscription values, the first subscription value corresponding to a first set of transportation criteria and the second subscription value corresponding to a second set of transportation criteria;
• increasing a travel cost value associated with the given travel service in response to determining that the likelihood of consumption value for the first user is less than the threshold, the travel cost value being based on at least one of the first or second maximum purchase amounts;
• and
• generating for display in a graphical user interface to the first user, an interactive visual representation of the given travel service that represents the travel cost value, the given travel service being associated with a cost that is greater than the first maximum purchase amount and less than the second maximum purchase amount.